docker - "Kind"写入节点创建集群失败

Question

我正在尝试为我的 kubernetes 集群设置一个 kind 集群。不幸的是，它在编写节点时准备好节点后失败了。我将附上输出和一些信息。提前感谢您的帮助!

干杯

错误

$ kind create cluster --config kind-config.yaml 

Creating cluster "kind" ...
 ✓ Ensuring node image (kindest/node:v1.20.2) 🖼 
 ✓ Preparing nodes 📦 📦 📦 📦  
 ✗ Writing configuration 📜 
ERROR: failed to create cluster: failed to generate kubeadm config content: failed to get kubernetes version from node: failed to get file: command "docker exec --privileged kind-worker3 cat /kind/version" failed with error: exit status 1
Command Output: Error response from daemon: Container c41566958be2239a9470ef2ea636c4b21958ee7620086f526954a02e4a605106 is not running

类配置yaml

apiVersion: kind.x-k8s.io/v1alpha4
kind: Cluster
nodes:
  - role: control-plane
  - role: worker
  - role: worker
  - role: worker

我的节点

$ kubectl get nodes -o wide

NAME      STATUS   ROLES                  AGE     VERSION   INTERNAL-IP     EXTERNAL-IP   OS-IMAGE       KERNEL-VERSION     CONTAINER-RUNTIME
gemini    Ready    control-plane,master   3d18h   v1.20.2   192.168.2.203   <none>        Ubuntu 20.10   5.8.0-1015-raspi   docker://19.3.13
phoenix   Ready    <none>                 3d17h   v1.20.2   192.168.2.129   <none>        Ubuntu 20.10   5.8.0-1015-raspi   docker://19.3.13
taurus    Ready    <none>                 3d17h   v1.20.2   192.168.2.201   <none>        Ubuntu 20.10   5.8.0-1015-raspi   docker://19.3.13
virgo     Ready    <none>                 3d17h   v1.20.2   192.168.2.202   <none>        Ubuntu 20.10   5.8.0-1015-raspi   docker://19.3.13

我的集群上正在运行什么

$ kubectl get all --all-namespaces

NAMESPACE              NAME                                             READY   STATUS    RESTARTS   AGE
default                pod/nginx-6799fc88d8-62cjd                       1/1     Running   1          18h
kube-system            pod/calico-kube-controllers-86bddfcff-ccrhg      1/1     Running   7          3d18h
kube-system            pod/calico-node-jddnl                            1/1     Running   4          3d17h
kube-system            pod/calico-node-nxwlw                            0/1     Running   7          3d18h
kube-system            pod/calico-node-stnzs                            0/1     Running   0          52s
kube-system            pod/calico-node-zrxzl                            1/1     Running   4          3d17h
kube-system            pod/coredns-74ff55c5b-kb2nm                      1/1     Running   7          3d18h
kube-system            pod/coredns-74ff55c5b-wsgs5                      1/1     Running   7          3d18h
kube-system            pod/etcd-gemini                                  1/1     Running   8          3d18h
kube-system            pod/kube-apiserver-gemini                        1/1     Running   8          3d18h
kube-system            pod/kube-controller-manager-gemini               1/1     Running   11         3d18h
kube-system            pod/kube-proxy-7fcjz                             1/1     Running   8          3d18h
kube-system            pod/kube-proxy-84rr7                             1/1     Running   4          3d17h
kube-system            pod/kube-proxy-lc88w                             1/1     Running   4          3d17h
kube-system            pod/kube-proxy-v4qd9                             1/1     Running   4          3d17h
kube-system            pod/kube-scheduler-gemini                        1/1     Running   9          3d18h
kubernetes-dashboard   pod/dashboard-metrics-scraper-79c5968bdc-mlb4s   1/1     Running   7          3d18h
kubernetes-dashboard   pod/kubernetes-dashboard-7448ffc97b-nq5c9        1/1     Running   7          3d18h

NAMESPACE              NAME                                TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)                  AGE
default                service/kubernetes                  ClusterIP   10.96.0.1        <none>        443/TCP                  41h
kube-system            service/calico-etcd                 ClusterIP   10.96.232.136    <none>        6666/TCP                 3d18h
kube-system            service/calico-typha                ClusterIP   10.109.108.233   <none>        5473/TCP                 3d18h
kube-system            service/kube-dns                    ClusterIP   10.96.0.10       <none>        53/UDP,53/TCP,9153/TCP   3d18h
kubernetes-dashboard   service/dashboard-metrics-scraper   ClusterIP   10.110.70.52     <none>        8000/TCP                 3d18h
kubernetes-dashboard   service/kubernetes-dashboard        NodePort    10.106.194.127   <none>        443:31741/TCP            3d18h

NAMESPACE     NAME                         DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR            AGE
kube-system   daemonset.apps/calico-node   4         4         2       4            2           kubernetes.io/os=linux   3d18h
kube-system   daemonset.apps/kube-proxy    4         4         4       4            4           kubernetes.io/os=linux   3d18h

NAMESPACE              NAME                                        READY   UP-TO-DATE   AVAILABLE   AGE
default                deployment.apps/nginx                       1/1     1            1           18h
kube-system            deployment.apps/calico-kube-controllers     1/1     1            1           3d18h
kube-system            deployment.apps/coredns                     2/2     2            2           3d18h
kubernetes-dashboard   deployment.apps/dashboard-metrics-scraper   1/1     1            1           3d18h
kubernetes-dashboard   deployment.apps/kubernetes-dashboard        1/1     1            1           3d18h

NAMESPACE              NAME                                                   DESIRED   CURRENT   READY   AGE
default                replicaset.apps/nginx-6799fc88d8                       1         1         1       18h
kube-system            replicaset.apps/calico-kube-controllers-56b44cd6d5     0         0         0       3d18h
kube-system            replicaset.apps/calico-kube-controllers-86bddfcff      1         1         1       3d18h
kube-system            replicaset.apps/coredns-74ff55c5b                      2         2         2       3d18h
kubernetes-dashboard   replicaset.apps/dashboard-metrics-scraper-79c5968bdc   1         1         1       3d18h
kubernetes-dashboard   replicaset.apps/kubernetes-dashboard-7448ffc97b        1         1         1       3d18h

Answer 1

答案不具体，从一开始就做好准备。有一个巨大的封闭Cannot create cluster due to docker exec cat /kind/version failing github 问题没有解决，但是..

您遇到的问题范围很广，根本原因可能完全不同。

BenTheElder - kind creator/maintainer :

This part: Command Output: Error response from daemon: Containerf2a2d9c8f9c2eca9aeec7f10249eb205b02c8a5f41e5bf1145b5a8e4b63da123 isnot running

That tells us that the node container is not running. That eithermeans the entrypoint failed or your host killed it, both either due tosome obscure bug we haven't found yet, or more likely an issue withyour host environment.

Please file your own issue with much more details. This issue isnon-specific and has discussed many different problems, as outlinedabove.

所以请在 Github 上创建新问题，很可能那个地方现在最适合解决此类问题。

我还发现您的问题可能来自 Docker Installed with Snap .参见 https://github.com/kubernetes-sigs/kind/issues/1288#issuecomment-631673479 .来自 SNAP 的 Docker 知道与 Kind 一起工作的问题，而且它实际上并没有得到 KIND 团队的支持

snap is in the known-issue document, the snap docker package has anumber of issues, e.g. no access to temp directories. I don'trecommend snap for docker and we don't really support this.

---

A small note: we've worked around most of the snap issues for now ifyou're just managing clusters, but I still don't recommend snap fordocker.

If you're seeing an issue similar to this, it means the node containerexited early for some reason. That usually means the host environmentis broken, but occasionally has meant we need to work-around e.g. lesscommon filesystems with device mapper issues.

Please attempt to capture node logs with kind create cluster --retain,kind export logs, and file an issue with the logs uploaded. We'll tryto identify the cause based on these.

EDIT: to be extra clear: "Cannot create cluster due to docker exec cat/kind/version failing" is a symptom, please file an issue with thedetails on your specific failure so we can identify the actual rootcause if you encounter this.

This issue is locked because it got off-topic from the original rootcause and kept being used for new problems that just happen to havethe same symptom. This symptom is common for edge cases with the nodessuddenly terminating very early on because it's one of the firstactions we take against the running node.